If a catheter approach to the treatment of atrial fibrillation (Afib) can be developed, the clinical implications for patients suffering from Afib may be enormous. One of the fundamental questions concerning the concerning the ablation of Afib is whether a full set of "Maze" like linear lesions are necessary to effectively ablate every case of human Afib. We have developed a safe and efficient catheter technology that is capable of obtaining intracardiac electrograms and producing long continuous transmural lesions within the atria. Preliminary data shows that this catheter is reliable, safe and can be used to ablate acute and chronic Afib in the dog model. The primary aim of this proposal is to use this unique technology to determine whether it is possible to ablate chronic Afib with minimal lesions using a mapping-directed approach to lesion placement instead of simply generating a full set of predetermined anatomically-placed lesions, as used in surgical approaches. We will direct our effort toward proving or disproving the following hypotheses: 1) Highly fractionated stable atrial electrical activity identifies the tissues which drive and perpetuate Afib activity 2) linear lesions placed at these locations ablate chronic Afib. 3) This method of Afib ablation is as effective as anatomically-paced lesions. To address these issues we will investigate the placement od linear lesions in regions of the atria with the greatest fractionated activity, as determined via an intracardiac multi-electrode mapping system. We propose to investigate 2 groups of dogs with chronic Afib. Group #1)20 dogs who receive a complete set of LA and RA lesions whose placement is based purely on predefined anatomical locations. Group #2) 20 dogs in whom linear lesion placement is based solely on the results of atrial mapping. We will compare the acute and chronic conversion to sinus rhythm in these two groups. In addition we propose to further evaluate the safety and efficacy of this approach and to attempt t provide further clarification of the mechanisms of Afib ablation in this model by: 1) Characterizing the change in atrial activation before and after creation of linear lesions, 2) We will specifically try to address the changes in activation that may explain the mechanism of Afib conversion to flutter and sinus rhythm. 2) Evaluating sinus and AV nodal function and conduction times (intra-atrial and atrio-ventricular) pre and post ablation. 3) Assessing atrial mechanical function before and after linear lesion generation 4) Testing acute and Chronic ablation results by attempting to reinduce Afib throughout the 2 weeks post-ablation.